The development of methods for detecting and sequencing nucleic acids is critical to the diagnosis of genetic, bacterial, and viral diseases. See Mansfield, E. S. et al. Molecular and Cellular Probes, 9, 145–156 (1995). At present, there are a variety of methods used for detecting specific nucleic acid sequences. Id. However, these methods are complicated, time-consuming and/or require the use of specialized and expensive equipment. A simple, fast method of detecting nucleic acids which does not require the use of such equipment would clearly be desirable.
Colloidal gold-protein probes have found wide applications in immunocytochemistry [S. Garzon and M. Bendayan, “Colloidal Gold Probe: An Overview of its Applications in Viral Cytochemistry,” in “Immuno-Gold Electron Microscopy,” Ed. A. D. Hyatt and B. T. Eaton, CrC Press, Ann Arbor, Mich. (1993); J. E. Beesley, Colloidal Gold: A New Perspective for Cytochemical marking,” Oxford University Press, Oxford, (1989)]. These probes have been prepared by adsorbing the antibodies onto the gold surface from an aqueous solution under carefully defined conditions. The complexes produced in this manner are functional but suffer from several drawbacks: e.g., some of the protein desorbs on standing, liberating antibody into solution that competes with adsorbed antibodies for the antigen target; the activity is low since the amount adsorbed is low and some of the antibody denatures on adsorption; and the protein-coated particles are prone to self aggregation, especially in solutions of high ionic strength. An alternative means for preparing nanoparticle-protein probes has been described by J. E. Hainfeld, R. D. Leone, F. R. Furuya, and R. D. Powell (U.S. Pat. No. 5,521,289, May 28, 1996, “Small Organometallic Probes”). Typically, this procedure involves reduction of a gold salt in an organic solvent containing a triarylphosphine or mercapto-alkyl derivative bearing a reactive substituent, X, to give small nanoparticles (50–70 gold atoms) carrying X substituents on linkers bound to the surface through Au—P or Au—S bonds. Subsequently the colloidal solution is treated with a protein bearing a substituent Y that reacts with X to link the protein covalently to the nanoparticle. Work with these nanoparticle is limited by the poor water solubility of many proteins, which limits the range of protein-nanoparticle conjugates that can be utilized effectively. Also, since there are only a few gold atoms at the surface of these particles, the number of “capture” strands that can be bound to the surface of a given particle is very low.
A variety of methods have been developed for assembling metal and semiconductor colloids into nanomaterials. These methods have focused on the use of covalent linker molecules that possess functionalities at opposing ends with chemical affinities for the colloids of interest. One of the most successful approaches to date, Brust et al., Adv. Mater., 7, 795–797 (1995), involves the use of gold colloids and well-established thiol adsorption chemistry, Bain & Whitesides, Angew. Chem. Int. Ed. Engl., 28, 506–512 (1989) and Dubois & Nuzzo, Annu. Rev. Phys. Chem., 43, 437–464 (1992). In this approach, linear alkanedithiols are used as the particle linker molecules. The thiol groups at each end of the linker molecule covalently attach themselves to the colloidal particles to form aggregate structures. The drawbacks of this method are that the process is difficult to control and the assemblies are formed irreversibly. Methods for systematically controlling the assembly process are needed if the materials properties of these structures are to be exploited fully.
The potential utility of DNA for the preparation of biomaterials and in nanofabrication methods has been recognized. In this work, researchers have focused on using the sequence-specific molecular recognition properties of oligonucleotides to design impressive structures with well-defined geometric shapes and sizes. Shekhtman et al., New J. Chem., 17, 757–763 (1993); Shaw & Wang, Science, 260, 533–536 (1993); Chen et al., J. Am Chem. Soc., 111, 6402–6407 (1989); Chen & Seeman, Nature, 350, 631–633 (1991); Smith and Feigon, Nature, 356, 164–168 (1992); Wang et al., Biochem., 32, 1899–1904 (1993); Chen et al., Biochem., 33, 13540–13546 (1994); Marsh et al., Nucleic Acids Res., 23, 696–700 (1995); Mirkin, Annu. Review Biophys. Biomol. Struct., 23, 541–576 (1994); Wells, J. Biol. Chem., 263, 1095–1098 (1988); Wang et al., Biochem., 30, 5667–5674 (1991). However, the theory of producing DNA structures is well ahead of experimental confirmation. Seeman et al., New J. Chem., 17, 739–755 (1993).
A major challenge for those working in the area of DNA detection (G. H. Keller, M. M. Manak, DNA Probes (Stocktonton Press, New York, 1989)) involves the development of methods that do not rely on polymerase chain reaction (PCR) or comparable target amplification systems. Besides added cost, methods based on PCR require additional instrumentation, thermocycling, and reagents that are not ideal for point-of-care or field use. Another restrictive requirement of most DNA detection systems, regardless of their need for PCR, is a thermal stringency wash to achieve desired analyte selectivity. In 1997, an oligonucleotide-modified Au nanoparticles were introduced as colorimetric DNA detection probes (R. Elghanian, J. J. Storhoff, R. C. Mucic, R. L. Letsinger, C. A. Mirkin, Science 277, 1078–1081 (1997)). In contrast with other probes, such as fluorophores or radioactive labels, nanoparticles can be used in variety of detection formats that take advantage of their size dependent scattering, catalytic properties, absorption characteristics, and Raman-enhancing properties to develop systems that are substantially more sensitive and selective than their molecular fluorophore counterparts (J. J. Storhoff, R. Elghanian, R. C. Mucic, C. A. Mirkin, R. L. Letsinger, J. Am. Chem. Soc. 120, 1959–1964 (1998); R. A. Reynolds, C. A. Mirkin, R. L. Letsinger, J. Am. Chem. Soc. 122, 3795–3796 (2000); T. A. Taton, C. A. Mirkin, R. L. Letsinger, Science 289, 1757 (2000); T. A. Taton, L. Gang, C. A. Mirkin, J. Am. Chem. Soc. 123, 5164 (2001); L. He et al., J. Am. Chem. Soc. 122, 9071 (2000)). Moreover, gold particles that have been heavily functionalized with oligonucleotides exhibit extraordinarily sharp melting profiles that translate into higher target selectivities (Storhoff (1998); Reynolds (2000); Taton (2000)). Despite the extraordinary sensitivity and selectivity of these systems, all rely on thermal stringency washes to differentiate target strands from ones with mismatches. Accordingly, there is a need for DNA detection methods that do not rely on thermal stringency washes to distinguish base mismatches.